Conventionally, monitors enabling an observer to view two-parallax images captured from two viewpoints stereoscopically by using a specific device, such as a pair of stereoscopic vision glasses, have been in practical use. Furthermore, in recent years, monitors enabling an observer to view multi-parallax images (e.g., nine-parallax images) captured from a plurality of viewpoints stereoscopically with the naked eyes by using a beam control element, such as a lenticular lens, have also been in practical use. Such two-parallax images and nine-parallax images displayed on monitors enabling stereoscopic vision may be generated by estimating depth information of an image captured from one viewpoint and performing image processing with the information thus estimated.
As for medical image diagnosis apparatuses, such as X-ray computed tomography (CT) apparatuses, magnetic resonance imaging (MRI) apparatuses, and ultrasound diagnosis apparatuses, apparatuses capable of generating three-dimensional medical image data (hereinafter, referred to as volume data) have been in practical use. Conventionally, volume data generated by such a medical image diagnosis apparatus is converted into a two-dimensional image by various types of image processing, and is displayed two-dimensionally on a general-purpose monitor. For example, volume data generated by a medical image diagnosis apparatus is converted into a two-dimensional image that reflects three-dimensional information by volume rendering processing, and is displayed two-dimensionally on a general-purpose monitor.
In the conventional techniques, however, it is difficult to display the three-dimensional images capable of providing a stereoscopic view and the two-dimensional images in a manner suitable for each of different circumstances.